Is there an end in sight to the 鈥渉omework wars?鈥�
Homework is one those never-ending debates in K-12 circles that re-emerges every few years, bringing with it a new collection of headlines. Usually they bemoan how much homework students have, or highlight .
Now, a new analysis from the Center for American Progress suggests a more fruitful way of thinking about this problem. Maybe, it suggests, what we should be doing is looking at what students are routinely being asked to do in take-home assignments, how well that homework supports their learning goals (or doesn鈥檛), and make changes from there.
The analysis of nearly 200 pieces of homework concludes that much of what students are asked to do aligns to the Common Core State Standards鈥攁 testament to how pervasive the standards are in the U.S. education system, even though many states have tweaked, renamed, or replaced them. However, most of the homework embodied basic, procedural components of the standards, rather than the more difficult skills鈥攕uch as analyzing or extending their knowledge to new problems.
鈥淲e were surprised by the degree of alignment. And we were also surprised by the degree that the homework was rote, and how much some of this stuff felt like Sudoku,鈥� said Ulrich Boser, a senior fellow at CAP. 鈥淚t made the homework debate make a lot more sense about why parents are frustrated.鈥�
It is also similar to the findings of groups like the Education Trust,
Collecting Homework Samples
The CAP analysis appears to be one of the first studies to look at homework rigor using a national survey lens. Many studies of homework are based on one school or one district鈥檚 assignments, which obviously limits their applicability. Attempts to synthesize all this research have led to some hard-to-parse conclusions. concludes there鈥檚 some connection for grades 6-12 between homework and test scores, but less so for elementary students, and less of an impact on actual grades.
Another problem is that students鈥� experiences with homework seem to vary so dramatically: A Brookings Institution report based on survey data from the National Assessment of Educational Progress concluded that, while on average students aren鈥檛 overburdened by homework,
The CAP analysis, instead, was based on getting a sample of parents from across the country to send in examples of their children鈥檚 homework. The researchers used MTurk, a crowdsourcing service offered by Amazon.com to recruit parents. Of the 372 parents who responded, the researchers got a pile of 187 useable assignments. Next, John Smithson, an emeritus researcher at the University of Wisconsin-Madison, had teams grade them on a taxonomy looking at both the content and the 鈥渃ognitive demand,鈥� or difficulty, of the work. The index fell on a 1 to 10 scale, with a score 4 to 6 range considered as 鈥済ood鈥� alignment.
The results? On average, math assignments fell within this range, while the ELA ones were slightly weaker, in the 3 to 5 range.
But the real eye-opening graphic is this one, which shows that by far the assignments were mostly low-level.
This makes some logical sense when you think about it. Just as with teaching and testing, it is much easier to write homework assignments prioritizing basic arithmetic drills and fill-in-the-blank vocabulary words than ones that get students to 鈥減rove鈥� or 鈥済eneralize鈥� some tenet. (I suspect prepackaged curricula, too, probably lean more toward rote stuff than cognitively demanding exercises.)
Here鈥檚 another explanation: Many teachers believe homework should be for practicing known content, not learning something new. This is partially to help close the 鈥渉omework gap鈥� that surfaces because some students can access parent help or help via technology, while other students can鈥檛. It鈥檚 possible that teachers are purposefully giving lower-level work to their students to take home for this reason.
To be sure, Boser said, it鈥檚 not that all lower-level work is intrinsically bad: Memorization does have a place in learning. But assignments like color-in-the-blank and word searches are probably just a waste of students鈥� time. 鈥淗omework assignments,鈥� the study says, 鈥渟hould be thought-provoking.鈥�
Study Limitations
The study does come with some significant limitations, so you must use caution in discussing its results. The surveyed population differs from the population at large, overrepresenting mothers over fathers and parents of K-5 students, and underrepresenting black parents. Also, the majority of the assignments the parents sent in came from the elementary grades.
The report makes suggestions on how districts can strategically improve the quality of their homework, rather than deciding to chuck it out altogether.
One is to is to audit homework assignments to make sure they鈥檙e actually useful at building some of the more difficult skills. Another is to extend the 鈥渃urriculum revolution鈥� of the last decade, which has focused more attention on the quality and alignment of textbooks and materials, to homework. A third is to use appropriate technology so students can access out-of-school supports for challenging homework.